Corey Montella's profile - activity

There recently was a bug in rosbag that recorded topics from the same node into one. It sounds like you triggered that.

https://code.ros.org/trac/ros/ticket/3755

You should update ROS or more specifically ros_comm.

When I view bag files of recorded autonomous navigation, the global plan, and local plan seem to all get recorded under the same topic, even though they are published under three separate topics. This causes the plan to flash constantly between the three. How can I record these topic separately?

For example my topics are: /move_base/DWAPlannerROS/global_plan /move_base/DWAPlannerROS/local_plan

And when I play back the bag file, the only topic is: /move_base/DWAPlannerROS/global_plan

But the points from the local plan are saved in that topic as well as the global plan points.

I had this same problem after updating Ubuntu. I came across this related answer and tried the suggestion and it worked:

http://answers.ros.org/question/2109/rviz-in-ros-electric

Basically I did sudo rm /usr/lib/libGL.so* and it worked.

Negative average delay comes from the transform being timestamped in the future when sent or by network delays when running in simulation.

In simulation time is sent over the /clock topic and may arrive after transforms sent from a different node with the same time.

A few nodes stamp data in the future, most noteably AMCL. This is valid for AMCL and other similar nodes publishing a small correction transform which is expected to remain approximately static and never have inherent velocity. The value is that AMCL can update at a slow rate but does not prevent transforms through it's correction to be used at a high rate.

When I use static_transform_publisher to provide a static transform, the average delay is negative. My questions are: 1) Is this a problem? 2) If so, is there a reason it's doing this, and how do I fix it?

For example, when I run:

rosrun tf static_transform_publisher 0 0 0 0 0 0 /frame1 /frame2 100

tf_monitor shows:

RESULTS: for /frame1 to /frame2
Chain is: /frame1 -> /frame2
Net delay     avg = -0.0473741: max = 0.000169573

Frames:
Frame: /frame2 published by /static_transform_publisher_1301872460950178911 Average Delay: -0.0997857 Max Delay: 0

All Broadcasters:
Node: /static_transform_publisher_1301872460950178911 10.3418 Hz, Average Delay: -0.0997857 Max Delay: 0

Corey, This sounds like a bug in pcl. I suggest that you file a ticket https://code.ros.org/trac/ros-pkg/newticket?component=point_cloud_perception&type=defect&&point_cloud_perception or email pcl-users@code.ros.org

The statistical outlier removal object has a setNegative() method which will extract outliers instead of inliers. Simply pass this method 'true'.

For the passthrough filter, you can use setFilterLimitsNegative(). Pass this method 'true' as well to get the outliers.